*3.4. SENP Processes the SUMO C Terminus*

In the next step, we verified that SENP is indeed required for C-terminal processing of SUMO. We constructed plasmids for ectopic expression of SUMO either with an N-terminal 3×HA tag (Figure 4A) or with a C-terminal 3×HA tag (Figure 4B) and transfected them into *L. donovani* wild type and *L. donovani* SENP−/<sup>−</sup> cl.1. The cells were grown to mid-logarithmic density, collected by centrifugation and lysed in SDS sample buffer. Samples representing equal cell numbers of *L. donovani*, *L. donovani* SENP−/−, *L. donovani* (3×HA-SUMO), *L. donovani* SENP−/<sup>−</sup> (3×HA-SUMO), *L. donovani* (SUMO-3×HA) and *L. donovani* SENP−/<sup>−</sup> (SUMO-3×HA) were separated by SDS-PAGE, blotted and stained with an anti-HA antibody (Figure 4C). No unspecific HA tag staining was observed for wild type and the SENP−/<sup>−</sup> mutant. Ectopic expression of 3×HA-SUMO in the wild type background resulted in a band corresponding to 25 kD, not the expected 16 kD of the triple-HA-tagged SUMO. The aberrant migration of SUMO in SDS-PAGE has been described before [43] and explains the observed band. We also observe numerous bands of higher molecular mass. Their spacing and varying intensities does not reflect the incremental size increases expected of SUMO homoconjugates, but rather suggests HA-tagged, SUMOylated substrate proteins. In the SENP−/<sup>−</sup> background, expression of the same transgene resulted in a slightly larger band, presumably representing the monomeric, non-processed 3×HA-SUMO. No larger HA-tagged bands were detectable, indicating that unprocessed SUMO is incapable of being conjugated to itself or to target proteins.

**Figure 3.** Replacement of SENP. (**A**) Schematic representation of LdBPK\_262070 replacement using the CRISPR/Cas9 technology. *SUMO*-targeting sgRNAs (grey arrowheads) and replacement cassettes were PCR-amplified and transfected into *L. donovani* (Cas9/T7RNAP). (**B**,**C**) RT-qPCR of RNA from *L. donovani* wild type (WT), SENP−/<sup>−</sup> clones 1–4, SENP−/<sup>−</sup> cl.1[pCLN-SENP], and *L. donovani* (Cas9/T7RNAP). (**B**) SENP-specific RT-qPCR. (**C**) Cas9-specific RT-qPCR. *n* = 2. (**D**) Whole genome sequencing of *L. donovani* wild type (WT), *L. donovani* (Cas9/T7RNAP), SENP−/<sup>−</sup> cl.1 and SENP−/<sup>−</sup> cl.2. Sequence reads were aligned to *L. donovani* chromosome 26. The ruler shows the position of the SENP CDS; the numbers refer to the position within chromosome 26. Read coverage is shown in blue.

− −

− − − −

− − **Figure 4.** SUMO processing by SENP. (**A**) Schematic drawing of pCL2N-3×HA-SUMO, a plasmid for ectopic expression of SUMO with an N-terminal triple HA tag. (**B**) Schematic drawing of pCL2N-SUMO-3×HA, a plasmid for ectopic expression of SUMO with a C-terminal triple HA tag. (**C**) Western blot of *L. donovani* wild type or SENP−/<sup>−</sup> null mutants, expressing 3×HA-SUMO or SUMO-3×HA, probed with anti-HA tag antibodies. *n* = 2. (**D**) Coomassie brilliant blue (CBB) staining of replicate SDS-PAGE gel, serving as a loading control. The positions and masses of protein size markers are indicated on the left. Original Western blot and gel images can be seen in Figure S1.

No HA-tagged proteins are visible when the C-terminally tagged SUMO-3×HA is expressed in the wild type background. Expression of the same chimera in SENP−/<sup>−</sup> cells, by contrast, yields HA-tagged SUMO. This demonstrates that C-terminal processing of SUMO depends on SENP. We conclude that SENP is required for processing and conjugation of SUMO to itself and/or to other proteins, and establishes C-terminal cleavage of SUMO as a critical step for SUMOylation in *Leishmania*.
